Easy Threading Job. Or?

[Dr. Rob]

Looking for ideas on a threading job that came in last week from one of the larger industrial suppliers of fasteners. A one-off order. Looking for the fastest, easiest, smoothest way to get done quick.

These are 200 x of cylindrical steel nuts, threaded internally M6 x 15 mm long (thru) that need to be given an external thread of M10, all the way. Diameter is now 9.9 mm. No critical tolerances to worry about, just a functional M10.

No prob, right? Just make an external M6 spigot to hold the nuts on while in the lathe chuck, and use a tailstock die holder under power.

Wrong. The torque from the die just turns everthing in the chuck. So, made it out of hex instead. No good. The torque from the die just torqued the M6 spigot right off.

Oookay- maybe single-point cut the thread, and just use the die to clean up? Nope- Tried cutting the thread all in one go, and the result was not good. (infeed 1.5 mm) So, two cuts then. Nope, still no good. Three cuts?

No, this is getting out of hand- this is a 200 x job. Three cuts per, means six hundred stops, goes, revs and infeeds and die chases. NO WAY. NO. Will NOT do. I want this done in the easiest fastest way.

Thought about using those three-pointed (well, 2آ½-pointed then) carbide inserts, thought about making a 60* multi-cutter to drive in the toolpost grinder, or an old tap, thought about using an old die cut in half as a one-pass tool...

Seriously, this is the esiest job I've had in years. Question is, what is the most effective way to go about it?

[John Stevenson]

Rob, Scrap what you have, buy a length of M10 allthread and drill and tap the M6, then part off.
Or part off to length and hold in a split threaded collet or bush and D&T the M6.

John S.

[Dr. Rob]

John!! Welcome back! Been thinking of you recently. We gotta talk gears later.

Was onto the allthread thing too, but it would mean (probably) having to deburr each piece twice. (each end). So, maybe your second alternative there. But, that's a fair amount of centering, drilling and tapping and parting. Was hoping for a two or three-move sequence.

On the other hand, I just came across a small CNC lathe for not too much money. Maybe this job is a good excuse to get into CNC!

[Dr. Rob]

On second thought, John, maybe you're right. If I go buy one of the combined drill & chipfeeding taps, I could get a pretty good pace going. Assuming they were parted off beforehand, then.

[Tony]

Dr.R --
since you've already listed just about any option i could think of off hand.. and johnS came up with a pretty good candidate, i'll just throw this out there..

since the ext thread isnt "critical" and only needs to be functional.. perhaps you could get away with cutting them all, 10 or 20 pcs at a time. thread them onto an M6 bolt (hex head) -- even some allthread with a nut tackwelded as a head -- run each one on pretty quick, so they 'jam up' against eachother.

take a very light skim cut with sharp HSS tool.. maybe even a few passes with a good file, to even out any bumps.. and also tighten up any play.

have at it with a single point.. two passes min, i'd guess. then a few passes with an M10 threadfile for cleanup.

not pretty i'll admit. but i'm sure they would pass for functional.

i've done this with ext. threaded bushings. they were a bit longer, and were brass. but it saved alot of time parting/facing individual pieces.

if the individual cylindrical nuts are provided already to length and have decent face finishes, i'd try to take advantage of that before parting and turning from scratch.

let us know how you work this out.

-knucklehead

[DR]

Regarding single point threading...

The 10 mm is roughly equivalent to a 3/8-16. For single pointing this thread on the CNC we'd probably take about 5 passes.

With that chip load these could most likely be held on a threaded mandrel. Too any passes for a manual operation on 200 pieces, though.

[Peter S]

Dr Rob,
Before you write off the external spigot idea, try using an Unbrako M6 capscrew (12.7 grade, not one of those crappy other brand 8.8 type). Use a bit of hex stock in the chuck, maybe drill the hex 6mm and weld head of capscrew to back end.
May work...

[darryl]

Your idea of the 60 deg. multi-cutter seems ok. I have tried it, and it worked in brass, though the homemade cutter didn't really cut it. It seems that with a rigid enough tool post spindle it should allow you to do the thread cutting in one pass. You would need to be able to adjust the left-right tilt angle of the toolpost spindle to match the angle of the thread as it would be once cut. You should also be able to do several nuts at a time on a good grade threaded steel mandrel, supported at the tailstock end with a live center. What would scrap the idea for me is having to make the cutter. Maybe that's available?

[Dr. Rob]

Cool, darryl. Did you make the cutter yourself, you said? Do you think using a thick slitting saw, ground to 60* might be a good start? How big were your threads?

This trick isn't new. It was (is?) commonly employed for large threads. I just haven't needed it before, and don't have the proper stuff for it, so I thought maybe there was a simplified trick for it. Like a tap running backwards @ just below center height, and with helix angle compensation. I can imagine that when bugs worked out and chips are flying, the work rate is astonishing.

Naw, scratch that last thing; I don't have the synchro stuff anyway.

[This message has been edited by Dr. Rob (edited 10-13-2003).]

[darryl]

Dr Rob, I made the cutter from a 1/16 inch slitting saw, 2 inch diameter. I ground each side to leave 60 degree teeth, but had to grind each side of each tooth for relief angles. It's not very consistent, and took time. I made the toolpost spindle with adjustable tilt to get the helix angle setting, and turned the lathe spindle by hand to bring the cutter across the workpiece, threading gears engaged. I don't recall the actual thread I was trying to create, but I think it was on a diameter of 3/4 in, probably a common coarse thread for that diameter. Aside from the cutter not being very good, I think there wasn't enough rigidity in my setup to control the cutting action. The teeth digging in caused lots of vibration instead of just chewing out chips. I think a stronger setup would have produced good results. In operation, the thread groove is created full depth, one pass. Another problem I had was the toolpost spindle needed torgue behind it, and instead is has speed. That's normal for grinding, but this is essentially a hogging out operation, where slower speed and higher torgue is called for. If you can supply that to a tiltable spindle, rigidly mounted to the crosslide, I'm sure it would work.
Hmm, if there's some way of using the cutter in a mill, and feeding the spindle down whilst rotating the workpiece- let me think on this. Would still need helix angle compensation. This is sounding more like milling a worm gear. Just brainstorming.

[This message has been edited by darryl (edited 10-13-2003).]

[Weston Bye]

Why not try roughing out the threads with a tool post grinder and thin wheel. The wheel would have to be tilted to provide the helix angle. With proper spacing you might even fit two wheels on the grinder to minimize the number of passes. Just adding ideas.

Wes

[Dr. Rob]

A good one too, wes. I took a look at that a while ago, with a super-thin cutoff wheel. It was still a tad too thick. Maybe I can get away with a Dremel wheel, but it'll take so long that the accumulated heat would almost heat my shop all winter long.

That was kind of a side track, though, since I was messing with the toolpost grinders to fiddle with the milling / hogging idea. Was concerned about the speed, and came back just in time to read darryl's done-it-before post, which pretty much closed that avenue for a while.

You know, I think this job really requires a new VersaMill or CNC machine. Hmmm, dub-dub-dub, e-bay dot....

Seriously, I really like that milling idea. Read up on it in my 1930's machining books. Great stuff. They were using both single and multiple (stacked?) milling cutters. Makes me want to build a serious rig for this purpose, but for this job it isn't worth it. Be fun though.

[DR]

After reading all the replies and your comments.....

What's the time frame on a job like this?

Do you have to bid it?

I built up my business years ago doing work for the fastener industry, fast turnaround and they didn't want to spend any money. Because of these reasons I've shied away from that business in the past few years.

With my current equipment I'd do the job by starting from scratch in a bar fed machine. I must have done hundreds of these kind of jobs where holding the part was a problem so starting over was the best solution.

I realize you may not have this kind of machinery, I'm more curious why a fastener house would send this job to you in the first place. Don't they know what equipment you have available?

[Dr. Rob]

Sure DR. Fair enough question.

In order, then:
I don't know exactly. It was a few weeks a couple of weeks ago, so I guess there is a week or two left.

No, didn't bid it. Over here, we don't worry about money the way Americans do. It costs what it costs. They didn't ask, and I didn't say. The emphasis is on getting the job done. Its a matter of honor and square business that they will pay, and that my price will be agreeable to all.

(that goes even further. I happen to know that the end customer is a construction company, who in turn of course has their customer who will also pay whatever it costs.)

Why? I'm a customer there. Go there once a week for ten years, and always buy onesies and twosies of oddball stuff. They know that I'm a one-off shop, and send me their customers that need special stuff. As it happens, the proprietor lives just down the street a couple hundred yards.

Equipment? Well, I have four lathes, two mills, a couple-three welders, gear hobber, grinders, pillar drills, plenty taps & dies...No, no problem there.

And, seeing as how the original idea was to just use an M10 die at one pass, it wouldn't seem as though I was lacking anything. Now really, you couldn't have expected the mandrel to torque right off, could you?

The methodology discussed here is in a way, only an exercise. Just looking at the most creative, pseudoscientific way of making a boring task into a masterpiece of effectivity.

There have been a lot of good ideas here. Maybe bundling them up ten at a time on all thread (x-tra strong!) and going at it with a top-of-the-line chip-feeding die will work just fine.

But, that wouldn't be playing the game!

Understand your other thoughts about the fastener industry. It's amazing that something so complex as a nut, could be so cheap & quick to produce at the nut factory. There is just no way to compete.

[Mcostello]

Why not use 1 adjustable die to rough out the threads, and split up the torque required. Then finish with another set to size. "Stacked" on a mandrel, time would not be increased that much if 2 passes would work.